Link to this post | posted 12 May, 2021 20:30

c.sunnen

So… I'm encountering this exact issue with J cluster Beem. I see the homing endonuclease in the C-terminal region of gp11 (terminase, large subunit) and the HHPred hits to large terminase in the N-terminal. I've also got gp1 hitting to large terminase (and T4's gp17, etc.) and gp9 (immediately upstream of gp11) hitting to small terminase. So. What should I actually call these genes? Here's what I'm thinking: gp1 = NKF; add to notes that "similarity to T4 GP17, DNA packaging protein" gp9 = terminase, small subunit gp11 = terminase, large subunit; add to notes "contains intein" Thoughts or suggestions? Nikki

Link to this post | posted 11 Nov, 2020 15:35

c.sunnen

Hi Wasique! Thanks for providing the entire annotation notebook entry! It's super helpful for making a call. While orphams are often not real, and you should definitely be skeptical, I would probably keep this gene exactly as you annotated it. First, it has really good coding potential (at least in the GM-self output). Second, it has a nice 1bp overlap with the previous gene (with the start you've chosen) and creates another 1bp overlap with the downstream gene; if you delete it, you create a big gap. These nice 1bp overlaps suggest an operon, and that all 3 genes (37, 38, 39) all get translated by the ribosome. Finally, it appears from your Phamerator screenshot that this specific area of the genome differs from closely related phages, as they have space for a gene (pham 7983, 6 members) that your genome does not; in fact, it looks like there may have been a deletion of some kind. So it's not surprising that you have a never-before-seen gene here; it's not like other phages in this cluster! Good job finding all the evidence. I think the evidence supports keeping it. Nikki Sunnen USciences PHAGES faculty SMART member

Link to this post | posted 10 Nov, 2020 16:55

c.sunnen

Hi! While annotating cluster J phage Beem, I came across a spot where I could either insert a gene or pick a further upstream start site (it is reverse) to minimize the gap. This is gene 71 in Phamerator. The evidence supporting picking a further upstream start site is the fact that it has a lot of 1:1 hits when blasted, and it includes all the coding potential. However, inserting a gene between 71 and 72 is supported by the fact that it would allow for a 4 bp overlap (the gene is surrounded by genes with 4 bp overlaps). Also, it has multiple 1:1 hits, especially with phages (such as NihilNomen) that have been very similar to Beem, as noted throughout the annotation process. But, inserting the gene would cut off about 75 bp of coding potential for gene 71. Which piece of evidence should be weighted more heavily (coding potential or 4 bp overlap)? Thank you! (posted on behalf of) Christina Scanlon, a freshman at the University of the Sciences

Link to this post | posted 11 Sep, 2020 18:52
c.sunnen	Yikes. This looks like it lost something it needs to run, as if something is wrong in the path. Why? Who knows. In my experience, this usually means a new installation of DNA Master, and uninstall/delete of the previous version.

Link to this post | posted 06 Aug, 2020 19:57

c.sunnen

Hi! In preparation for running some investigations beyond annotation this coming semester, we've been working though the suggestions in the 2013 "Exploring bacteriphage biology" document found on phagesdb, and we've been working through softberry to find terminators. We'd like to write up a "protocol" for our undergrads, but we're having a hard time finding info on how to interpret the data. Who else has used this program, and can you point us to pre-written protocols/resources? Most papers we found don't explain how they used it at all. Thanks! Nikki

Link to this post | posted 03 Jun, 2020 19:07

c.sunnen

Hi! While it makes no sense whatsoever, I'm getting some good hits in HHPred to "Caudovirus prohead serine protease" (98% probability, ~95% coverage; see attachment) for GC phage Rasputia gp130 (in phagesdb and Phamerator gp130. That's the same as "capsid maturation protease," no? Note that this is an orpham, and we've already got a capsid maturation protease (where it should be!) I'd like to call this simply "peptidase" for now, but I'd love a second opinion on this. It's such a good match in HHPred! 62Kb

Link to this post | posted 23 Mar, 2020 20:08
c.sunnen	Thanks, Debbie! I totally agree that it's not worth calling at this point, but boy is it an interesting case! As always, I appreciate the experts weighing in. Hope things are well!

Link to this post | posted 23 Mar, 2020 19:46

c.sunnen

Rasputia (GC cluster) appears to have 4 tRNAs in a cluster (see my attachment with screenshots from DNAMaster and tRNAScanSE). The one in question is what DNAMaster is calling feature 150. It appears to have an intron? It is way too long, and online Aragorn does not call it. BUT, tRNAScanSE DOES, and gives it a good infernal score (43. and a mostly cononical structure. However, it doesn't appear to have an anticodon. My hunch is that it's not real, but if we delete it, we'd be leaving a big hole. I know that we generally don't call it unless it's less than 90bp, but I'm really curious if this could be at all real. Unless I hear otherwise, we won't call it. But I'd love to know more if anyone can offer some insight. Thanks! Nikki 159Kb

Link to this post | posted 14 Nov, 2019 03:22

c.sunnen

Hi! My first suggestion is to digest longer; we typically do ours for at least an hour at 37deg or overnight at room temp. Good news is your DNA looks nice! Some smearing, but mostly intact. If you're uncertain if your enzyme is any good, you can try digesting some lambda DNA (provided you have some on hand). I hope this is helpful! Good luck! Nikki P.S. If you're using the protocol I've posted above, there have been a few edits, though it should still work in its current form.

Link to this post | posted 12 Jul, 2019 15:29

c.sunnen

I had something similar in Mashley (EG). Genes 42 and 43 were orphams, the result of a frameshift that split a gene in approximately half. Interestingly, they split a "membrane protein" into two "membrane protein" proteins! Both orphams retained several TM domains! So they both got called that way. In your case, I think calling them both NKF genes is probably appropriate (unless there is a function you can attribute to either/both combined). Without bench data to suggest otherwise, as long as it's got coding potential, it's probably safer to assume it does get made, even if it's non-functional.